/// GUIDES / MIGRATE & SYNC DATA

Migrate & sync data

The migration engine copies or syncs objects between any two buckets — across different accounts and across different clouds. Jobs run in the background, nothing is buffered on the server, and you watch progress live in the Transfers tab.

What the migration engine does

Point a source Fleet at a destination Fleet and storafleet moves the objects for you. Because every Fleet is just a connected bucket, the source and destination can live in separate accounts or on entirely different providers — the engine bridges them.

Run a migration

1. Pick the source Fleet and the destination Fleet.
2. Optionally filter by key prefix to move only part of the bucket.
3. Choose Copy (one-time) or Sync (mirror the source into the destination).
4. Choose conflict handling — skip, replace, or ask — for keys that already exist at the destination.
5. Start the job.

Background, resumable jobs

Migration jobs run in the background, so you can close the page and let them finish. There's no size limit and nothing is buffered on the server. Jobs are resumable, and you can watch live progress in the Transfers tab. When a job finishes you get a completion notification.

R2 fast-paths

When the destination is Cloudflare R2, there are two faster, provider-native options:

• Super Slurper — turbo bulk import into R2 for moving an existing bucket over in one large transfer.
• Sippy — copy-on-read live migration. R2 serves objects immediately and backfills the rest in the background, so there's zero downtime.

For more on how migrations work, see Migration, and for moving off AWS specifically see Migrate off AWS.